aboutsummaryrefslogtreecommitdiffstats
path: root/tools/video-tile-poster-test.js
Commit message (Collapse)AuthorAgeFilesLines
* A flick through the feed no longer drags background work through the middle ↵Adam Soltys11 days1-3/+3
| | | | | | | of it. "When I flick to scroll, it's not getting up to speed very quickly, and clicking around in the app feels a bit sluggish at times" (Adam, 2026-11-07). Measured on his Pixel 9 in the installed PWA: idle is quiet (0.16 s of script in 10 s), but three real flicks through Following spent 1.7 s of 9 on the main thread, frames up to 200 ms, and the work was the avatar thumbnailer finishing faces as they scrolled in: a synchronous canvas.toDataURL WebP encode per face (229 ms), the profile cache JSON.stringified twice per half-second flush (once just to measure it, up to ~2 MB with thumbnails), and a whole-page repaint per finished face that changes nothing on screen (feed rows keep the face they were admitted with). Now: the crop, resize and encode happen off the main thread (createImageBitmap with resizeWidth/Height, OffscreenCanvas.convertToBlob, FileReader), with the old canvas path as the fallback; a finished face is recorded once the page settles; scheduleRepaint holds while a finger is down or the page scrolled in the last 200 ms (at most 1.5 s), and the profile flush likewise (at most 5 s, forced on an account switch) and sizes the blob by summing rows instead of stringifying it. Tests: the create flow's heads-up / passphrase / check steps from earlier today broke every test that made a wallet by waiting for the words and clicking "Skip verification"; tools/puppeteer.js exports seedWords() and skipSeedCheck() and the 96 tests use them. room-faces and avatar-profile fail the same way on 8858591 (row.hover, scrollIntoView), before today's changes Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_01VMan6Ts5Tktega8VbLg31q
* Media grids: one tile for a clip uploaded twice, and a picture that won't ↵Adam Soltys2026-10-221-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | load stays out - Videos: one person's clips of one size to the byte are one clip. Beatbox Serenade posted the same video as a note and as a video event, each upload a fresh Blossom hash, so neither the file nor the words matched. Only a size the post or the host states counts (a bitrate x duration guess is marked sizeGuess and never matched). - Images: a picture whose load failed was hidden by a class the next repaint rewrote, so the broken-picture glyph came back (a deleted nostr.media file in #artstr). Failed pictures are now remembered for the session and left out of the grid. - A tile fetching its own frame froze the whole tile against repaints, so one built for the one-to-a-row layout never got its --ar in justified rows (zero width). Only the player and its stand-ins are frozen now; the render draws the cost badge itself. Tile pictures no longer size their tile (a frame-grab <video>'s 320px stretched its row). - Test fixtures that served one file under several names (or one imeta size for every clip) now differ per clip. Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_01K5xtiFrkFVsrkWx6JDQjiT
* Videos grid: a clip the player can't load leaves the grid instead of sitting ↵Adam Soltys2026-10-221-6/+30
| | | | | | | | | | | | | there as a blank tile A deleted file on Primal's blossom answers 404 without CORS headers, so the probe saw only a quiet host (not a 404), the server had no still for it, and the tile's fallback player failed with nothing listening: a blank box (Adam's iPhone). The player's error now hides the tile and marks the clip gone for a day, as a 404 the probe could read already did. Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_01K5xtiFrkFVsrkWx6JDQjiT
* Videos grid: a tile without a still of its own shows the server's (vmedia ↵Adam Soltys2026-10-091-0/+95
thumb.jpg) at once — about a second the first time anyone asks for that clip, a CDN hit after — instead of waiting on the browser's probe (64 KB) and frame grab (≈1 MB of the clip, decoded) per tile, which on a phone left most of the grid as empty boxes. The frame grab still runs for a clip the server had no picture for. A known-heavy clip wears its size badge over the still. vmedia makes 8 stills at once (was 4). tools/video-tile-poster-test.js Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_01Vps2H27rhbG9ZqseGfMrc3